Michael B. Merickel

Noninvasive medical imaging system and method for the identification and 3-D display of atherosclerosis and the like

There is disclosed an image processing, pattern recognition and computer graphics system and method for the noninvasive identification and evaluation of atheroscelerosis using multidimensional Magnetic Resonance Imaging (MRI). Functional information, such as plaque tissue type, is combined with structure information, represented by the 3-D vessel and plaque structure, into a single composite 3-D display. The system and method is performed with the application of unsupervised pattern recognition techniques and rapid 3-D display methods appropriate to the simultaneous display of multiple data classes. The results are a high information content display which aids in the diagnosis and analysis of the atherosclerotic disease process, and permits detailed and quantitative studies to assess the effectiveness of therapies, such as drug, exercise and dietary regimens.

Model-based boundary estimation of complex objects using hierarchical active surface templates

A method for the segmentation of complex, three-dimensional objects using hierarchical active surface templates is presented. The templates consist of one or more active surface models which are specified according to a priori knowledge about the expected shape and location of the desired object. This allows complex objects to be naturally modeled as collections of simple subparts which are geometrically constrained. The template is adaptively deformed by the three-dimensional image data in which it is initialized such that the template boundaries are brought into correspondence with the assumed image object. An external energy field is developed based on a vector distance transform such that the surfaces are deformed according to object shape. The method is demonstrated by the segmentation of the human brain from three-dimensional magnetic resonance images of the head given an a priori model of a normal brain.

Noninvasive quantitative evaluation of atherosclerosis using MRI and image analysis.

A new medical image analysis system to quantify atherosclerosis in the lower abdominal aorta using magnetic resonance imaging is described. This medical image analysis and display system permits the quantification of the three-dimensional (3D) properties of the vessel wall and lumen cross-sectional area and volumes. Preliminary results of employing this medical image analysis capability on magnetic resonance images demonstrated a twofold increase in wall volume per unit vessel length, corresponding to intimal thickening, before luminal narrowing was detected. This work demonstrated the feasibility and usefulness of quantitatively evaluating the 3D properties of the vessel lumen and wall by using a combination of magnetic resonance imaging and image analysis. The demonstration that intimal wall thickening is observed in images before observable occlusion of the lumen can be expected to provide an important early indicator of the future development of atherosclerosis. Such capability will permit detailed and quantitative studies to assess the effectiveness of therapies, such as drug, exercise, and dietary regimens.

3D reconstruction: the registration problem

Abstract Three dimensional (3D) reconstruction of serial sections through biological tissue is a very important method for visualizing and quantifying the 3D relationships between internal structures. This project involves the automation of the 3D reconstruction process, focusing on the problem of registering or aligning successive serial sections. Serial sections can be misregistered with respect to rotation, translation and scaling (if magnification is not constant). A linear least squares fitting procedure is developed to determine the required registration transformation based on a set of labeled feature points. The results obtained from using two different types of feature points are evaluated and compared: (1) manually selected feature points and (2) characteristic shape points computed from the contours which describe specific shape properties (e.g., centroid and radius weighted mean point). These results demonstrate that good registration accuracy is obtained using both kinds of feature points under conditions in which the contours do not have exact matches.

Breast lesion discrimination using statistical analysis and shape measures on magnetic resonance imagery

Magnetic resonance images of intact human breast tissue are evaluated using statistical measures and shape analysis. In this paper, the Mahalanobis distance measurement and a related F-statistical value demonstrate that breast lesions are statistically separable from normal breast tissue. The minimum set of parameters to provide first order statistical separability between fibroadenomas, cysts, and carcinomas are T1-weighted, T2-weighted, and Dixon opposed pulse sequences. Tumor shape is quantified by development of a compactness measure and a spatial frequency analysis of the lesion boundary. Malignant lesions are shown to be separable from benign lesions based on quantitative shape measures.

Supervising ISODATA with an information theoretic stopping rule

Abstract New biomedical imaging modalities, such as Magnetic Resonance Imaging (MRI), provide fertile multidimensional environments for the automatic identification of biological soft tissues, but lack the a priori information required to appropriately train supervised classifiers. Hierarchical cluster analysis techniques can generate this information but they are inefficient on large data sets. We have developed an unsupervised clustering method that is a variant of the well known ISODATA algorithm. We replaced the heuristic rules that control ISODATA with rules that search for the minimum value of an information theoretic criterion. The criteria investigated in this study are Akaikes Information Criterion (AIC) and the Consistent AIC (CAIC). Both measure the global fit of a cluster model to the input data, and the smallest criterion value suggests the best fit. We tested the new method on multivariate Gaussian and real world data, including MR images of normal and diseased tissue in vivo.

Assessment of the reliability of the determination of carotid artery lumen sizes by quantitative image processing of magnetic resonance angiograms and images

In order to use MR imaging to assess progression or regression of atherosclerosis, one must have an idea of the reproducibility of the imaging and image processing techniques. The ability of dark-blood MRI and semiautomated image processing to reproducibility measure the inner boundary of the carotid arteries was evaluated and compared with results obtained using bright-blood MRA. MRI and MRA images were obtained for two normal and two diseased volunteers six times each over a short period of time (6 months). The carotid bifurcation was used to align slices from different imaging sessions. The area for each vessel (right and left common, internal and external carotid artery) was determined for the six imaging sessions. The standard deviations of each lumen area normalized to the average area were computed for each vessel segment for each volunteer. For the common, internal, and external carotids, the averaged normalized standard deviations for MRI were 8, 12, and 17% and for MRA were 6, 8, and 13%. Lumen sizes obtained by MRI and MRA were found to be not statistically different. Eccentric plaques not seen on MRA were visualized by MRI. In conclusion, dark-blood MRI with semiautomated image processing yields reliable lumen areas that are in agreement with those obtained by MRA.

Segmentation of the brain from 3D MRI using a hierarchical active surface template

The accurate segmentation of the brain from three-dimensional medical imagery is important as the basis for visualization, morphometry, surgical planning and intraoperative navigation. The complex and variable nature of brain anatomy makes recognition of the brain boundaries a difficult problem and frustrates segmentation schemes based solely on local image features. We have developed a deformable surface model of the brain as a mechanism for utilizing a priori anatomical knowledge in the segmentation process. The active surface template uses an energy minimization scheme to find a globally consistent surface configuration given a set of potentially ambiguous image features. Solution of the entire 3D problem at once produces superior results to those achieved using a slice by slice approach. We have achieved good results with MR image volumes of both normal and abnormal subjects. Evaluation of the segmentation results has been performed using cadaver studies.

High-resolution proton NMR spectra of human arterial plaque

Well-resolved proton (1H) NMR spectra of solid human arterial plaque can be acquired. Studies have been carried out of human fatty plaque obtained postmortem (ex vivo), the total lipids extracted from human atheroma, and a model mixture of cholesteryl esters whose lipid composition resembles that of human atheroma. In each case, well-resolved 1H NMR spectra were obtained at body temperature (37 degrees C), with little or no underlying broad signal. Such sharp 1H NMR spectra are typical of isotropic fluids, whereas solid and liquid-crystalline materials give rise to much broader spectral lines. The results suggest the sharp 1H NMR spectra of human atheromatous lesions at body temperature are due largely to the presence of intracellular and extracellular droplets of cholesteryl esters in the isotropic liquid phase. These findings provide a necessary basis for use of 1H NMR techniques to image quantitatively the lipid constituents of human atheroma in vivo, and to study their chemical and physical properties.

Multispectral pattern recognition of MR imagery for the noninvasive analysis of atherosclerosis

An image processing, pattern recognition, and computer graphics system is described for the noninvasive identification and evaluation of atherosclerosis using multidimensional magnetic resonance imaging (MRI). Particular emphasis has been placed on the problem of developing a pattern recognition system for noninvasively identifying the different plaque classes involved in atherogenesis using minimal a priori information. A rapid 3-D display system is also described for the simultaneous display of multiple data classes resulting from the tissue identification process. The results of this work provide a high-information-content display, which will aid in the diagnosis and analysis of the atherosclerotic disease process, and permit detailed and quantitative studies to assess the effectiveness of therapies, such as drug, exercise, and dietary regimens.<<ETX>>